Hi LC, i did check re fixing bolts-these are not connected to the ground system of the smps.SMPS probably have GND connected to bolt pad and thus to chassis and that's not OK since loop is formed. Try to isolate both SMPS PCB fixing bolts and brake the loop.
Quan
OK, than, meaning SMPS PCB is completely isolated from chassis.
If you carefully look to stereo connection sch diagram in a separate A4 page included in the package, there is no 10 Ohm resistor anywhere near. Please connect exactly as it is proposed in stereo sch and there should be no noise present. A lot of First One modules has been built with no noise whatsoever.
I've ran into strange "issue" after changing my Magnepan 1.6 QR loudspeakers to a pair of "conventional" 3-way loudspeakers. After switching the amps on, there was a "pumping sound" and the woofers moved in and out.... Swapped my interconnect cables (the ones with the screen connected to the outer side of the cinch connector at one side where I had some answers about in the past, I didn't have issues with those cables with the Magnepans connected!) to another set of interconnects I have around and the problems completely dissapeared...
I have connected the grounding EXACTLY as recommended. (latest schematic)
I really liked the SQ of the interconnects I can't use anymore now....
Any idea why this behaviour started after swapping the loadspeakers??? (didn't touch anything else)
Thanks,
Mark.
I have connected the grounding EXACTLY as recommended. (latest schematic)
I really liked the SQ of the interconnects I can't use anymore now....
Any idea why this behaviour started after swapping the loadspeakers??? (didn't touch anything else)
Thanks,
Mark.
Hi Mark
Thanks for the report. What about swaping interconnect sides of where screen is connected to GND. Could be some EMI induced in the input cable (antenna), different load Z (speakers) could interact on specific HF. Could you try to add an extra serial L (0,5 uH) on the positive output wire, as such could prevent low Z at HF.
Also input HF reject cap to GND is only 100 pF. Adding parallel extra 100-220 pF would successfully kill HF RFI if this is the reason.
Few tips are always desirable, hopefully.
Thanks for the report. What about swaping interconnect sides of where screen is connected to GND. Could be some EMI induced in the input cable (antenna), different load Z (speakers) could interact on specific HF. Could you try to add an extra serial L (0,5 uH) on the positive output wire, as such could prevent low Z at HF.
Also input HF reject cap to GND is only 100 pF. Adding parallel extra 100-220 pF would successfully kill HF RFI if this is the reason.
Few tips are always desirable, hopefully.
Hi Andrej,
Thanks for your quick response.
I tried swapping interconnect sides, makes no difference...
As the amp responds positive to using different sort of interconnect, should I try extra capacitor parallel at the input first? (it's more easy for me to find 100 - 220 pF capacitor as finding high current 0,5 uH inductor)
Should I place this cap as close as possible to the amp pcb, or is it fine to fit it directly at the input connector? (the last option has my preference, as it has better access)
P.S. I assume both suggested solutions reduce the frequency range of the amp? Does this have any negative influence of the SQ? (I assume this isn't the case as the frequency range was VERY extended in the first place)
Mark.
Thanks for your quick response.
I tried swapping interconnect sides, makes no difference...
As the amp responds positive to using different sort of interconnect, should I try extra capacitor parallel at the input first? (it's more easy for me to find 100 - 220 pF capacitor as finding high current 0,5 uH inductor)
Should I place this cap as close as possible to the amp pcb, or is it fine to fit it directly at the input connector? (the last option has my preference, as it has better access)
P.S. I assume both suggested solutions reduce the frequency range of the amp? Does this have any negative influence of the SQ? (I assume this isn't the case as the frequency range was VERY extended in the first place)
Mark.
Last edited:
Yes, try input RC first. Present values enables very extensive freq response, additional 220 pF together with 470 Ohm (mandatory) input serial resistor would still measure 634 kHz/-3 dB, not so much affecting sq.
Last edited:
OK, than, meaning SMPS PCB is completely isolated from chassis.
If you carefully look to stereo connection sch diagram in a separate A4 page included in the package, there is no 10 Ohm resistor anywhere near. Please connect exactly as it is proposed in stereo sch and there should be no noise present. A lot of First One modules has been built with no noise whatsoever.
Did the grounding scheme as you suggested-noise free now
.Quan
Did the grounding scheme as you suggested-noise free now
Quan
That 's the way I like it.
It's paramount at First One that the input and output GND meets on PCB. If you look carefully GND wire from SMPS is exactly in between SPK- and input GND, meaning no voltage drop on cables or any resistance that could make a difference in comparison/subtraction of input/output voltage.
In R&D period I also tried this 10 Ohm but it was horrible, hum, some HF, you name it, so it was omitted.
Looking forward to hear your SQ opinion.
L.C.
Is it O.K. to fit the RC directly at the input connector? Am I correct in assuming: The series resistor comes first (direction from input terminal to amp) and after that comes the cap connected to ground?
What is prefered cap for this application? Film? I can get Wima FKP1 or 2.
Yes correct, first serial resistor from RCA cinch connector to PCB input, 220 pF between input pads on PCB.
Wima MKS 5-10%, 5 mm RM are perfect.
Grounding Scheme
Hi Guys,
As grounding has been discussed, I wonder if anyone has any opinions on my proposed grounding scheme. I am using 4 modules to 1 SMPS1200A400 power supply. This seems the simplest solution to me. I have made a power cable up that has four wires into each connector (positive negative and ground) so the power for each module comes from the same point.
I am trying to decide if I should connect the power supply ground to the chassis directly as shown, with a loop breaker or to not connect to chassis at all.
I have other amplifiers without a connection from audio ground to chassis ground and I don't suffer any problems from any RF interference.
What power rating should be used for the 1K resistor used in the back to back diode arrangement as LC suggests?
Thanks
Hi Guys,
As grounding has been discussed, I wonder if anyone has any opinions on my proposed grounding scheme. I am using 4 modules to 1 SMPS1200A400 power supply. This seems the simplest solution to me. I have made a power cable up that has four wires into each connector (positive negative and ground) so the power for each module comes from the same point.
I am trying to decide if I should connect the power supply ground to the chassis directly as shown, with a loop breaker or to not connect to chassis at all.
I have other amplifiers without a connection from audio ground to chassis ground and I don't suffer any problems from any RF interference.
What power rating should be used for the 1K resistor used in the back to back diode arrangement as LC suggests?
Thanks
Hi fluid
Proposed GND sch is OK. Ground-earth loop breaker could be installed to marked position without any problem. Yours power distribution wires are correctly made from SMPS connector ie. star like.
Resistor plus antiparallel diodes drops 0,55 V in either current direction so 1 k can be standard 0,25 W. It serves only to define SMPS's absolute GND potential according to earth, so these are close in terms of volts, also 1 k resistor presents low Z in relation to complete isolation.
The only problem I see is in SMPS's starting current capability, the question is will SMPS turn on four modules simultaneously. It has to be tested and this info will also be very beneficial to the others.
L.C.
Proposed GND sch is OK. Ground-earth loop breaker could be installed to marked position without any problem. Yours power distribution wires are correctly made from SMPS connector ie. star like.
Resistor plus antiparallel diodes drops 0,55 V in either current direction so 1 k can be standard 0,25 W. It serves only to define SMPS's absolute GND potential according to earth, so these are close in terms of volts, also 1 k resistor presents low Z in relation to complete isolation.
The only problem I see is in SMPS's starting current capability, the question is will SMPS turn on four modules simultaneously. It has to be tested and this info will also be very beneficial to the others.
L.C.
Thanks for the OK on the grounding scheme
I hope the SMPS will start 4 modules at normal bias
I wonder if there would be a way to simulate the bias current without using the modules to see if the SMPS will turn on?
I am still a little way off being able to test as I need to get the heatsinks polished and drilled for the modules first
I hope the SMPS will start 4 modules at normal bias
I wonder if there would be a way to simulate the bias current without using the modules to see if the SMPS will turn on?
I am still a little way off being able to test as I need to get the heatsinks polished and drilled for the modules first
Thanks for the OK on the grounding scheme
I hope the SMPS will start 4 modules at normal bias
I wonder if there would be a way to simulate the bias current without using the modules to see if the SMPS will turn on?
I am still a little way off being able to test as I need to get the heatsinks polished and drilled for the modules first
Quality takes time.
For that kind of simulation we would need exact model of SMPS1200 as well as First One.
So far we know SMPS1200 has very steep turn on characteristic, few ms from zero to Udc and First One at least 10 times slower current rise from zero to 0,28 A, times four ie. from zero to 1,12 A. At the end turn-on enable operation depends on current limit/shutdown curve of SMPS's current rise time sensing circuit. As we don't have an access to such data, the best way remains reality check.
I don't think I should have used the word "simulate".
I was thinking more of using a constant (variable) current dummy load to check at what load the SMPS would start up.
Something like this?
Variable Digital Load Kit
Or maybe it is just easier to use the First One Modules as test by varying bias current if SMPS won't turn on.
I was hoping to avoid using First One Modules as test subjects though
I was thinking more of using a constant (variable) current dummy load to check at what load the SMPS would start up.
Something like this?
Variable Digital Load Kit
Or maybe it is just easier to use the First One Modules as test by varying bias current if SMPS won't turn on.
I was hoping to avoid using First One Modules as test subjects though